Please login first
Hydroponics Monitoring Through UV-Vis Spectroscopy and Artificial Intelligence: Quantification of Nitrogen, Phosphorous and Potassium
* 1, 2 , 3 , 4 , 5 , 6 , 5 , 7 , 7, 8 , 2 , 2 , * 2
1  Faculty of Sciences of the University of Porto, Rua do Campo Alegre, s/n, 4169-007 Porto, Portugal
2  Centre for Applied Photonics, INESC TEC, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal
3  HIR Skåne AB, Borgeby Slottsväg 11, 237 91 Bjärred
4  RISE Food and Agriculture, Scheelevägen 17, 223 70 Lund, Sweden
5  Wageningen University and Research, Droevendaalsesteeg 1, 6708 PB WAGENINGEN, the Netherlands
6  Wageningen University and Research, Droevendaalsesteeg 1, 6708 PB Wageningen, the Netherlands
7  INESC TEC, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal
8  School of Science and Technology, University of Trás-os-Montes and Alto Douro, Vila Real, Portugal
Academic Editor: Manuel Algarra


In hydroponic cultivation, monitoring and quantification of nutrients is of paramount importance. Accurate, robust sensors for detection of Nitrogen, Phosphorus and Potassium (NPK) would be desired in horticultural production. Spectroscopy can be used for this, but other nutrients interfere and hinder accurate and reliable quantification.

In order to better understand and solve nutrients’ interferences, an orthogonal experimental design has been used, based on Hoagland fertilizer solutions, a widely used complete and complex nutrient mixture. The experimental factorial design consisted of eight orthogonal levels of N, P and K rendered on 83 of different samples of Hoagland solution, each one with its own specific concentration of NPK. Concentration ranges were varied in a target analyte independent style: [N]= [103.17-554.85] ppm; [P]= [15.06-515.35] ppm; [K]= [113.78-516.45] ppm, by dilution from individual stock solutions. This strategy allowed the variation of each parameter individually, maintaining the remaining constant, enabling the individual variations as well as their correlations to be obtained. A UV-Vis-based Artificial Intelligence-enhanced (AI) system was used for quantification of NPK on the analysed samples. It featured an advanced processing algorithm named Self-Learning Artificial Intelligence (SL-AI).

From the analysis of the acquired and processed data, it was possible to understand that N spectral features are dominant, whereas P and K will behave as interferents, with information on P properties not being very evident on spectra. The obtained results allowed very good quantifications for N and K, with errors of 6.7% (0.997) and 3.8% (0.987), respectively, to be achieved. Regarding P, as expected, only satisfactory results were obtained, corresponding to a qualitative grade. The developed system can be of great potential for monitoring and quantification of NPK in hydroponic platforms.


The authors would like to thank the financial support under the ERA-NET Cofund WaterWorks2015 Call, within the frame of the collaborative international consortium AGRINUPES. This ERA-NET is an integral part of the 2016 Joint Activities developed by the Water Challenges for a Changing World Joint Programme Initiative (Water JPI/002/2015).

RM acknowledges Fundação para a Ciência e Tecnologia (FCT) research contract grant (CEEIND/017801/2018).

AFS gratefully acknowledges the financial support provided by FCT (Portugal’s Foundation for Science and Technology) within grant (DFA/BD/9136/2020).

Keywords: Nitrogen; Phosphorus; Potassium; Hoagland; Optical Sensing; Artificial Intelligence